REVIEW 16 cited by
Large Wireless Model (LWM): A Foundation Model for Wireless Channels
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This paper presents Large Wireless Model (LWM) -- the world's first foundation model for wireless channels. Designed as a task-agnostic model, LWM generates universal, rich, contextualized channel embeddings (features) that potentially enhance performance across a wide range of downstream tasks in wireless communication and sensing systems. Towards this objective, LWM, which has a transformer-based architecture, was pre-trained in a self-supervised manner on large-scale wireless channel datasets. Our results show consistent improvements in downstream tasks when using the LWM embeddings compared to raw channel representations, especially in scenarios with high-complexity machine learning tasks and limited training datasets. This LWM's ability to learn from large-scale wireless data opens a promising direction for intelligent systems that can efficiently adapt to diverse tasks with limited data, paving the way for addressing key challenges in wireless communication and sensing systems.
Forward citations
Cited by 16 Pith papers
-
MultiPathFormer: Towards a Foundation Model for Multipath Wireless Propagation
Pretraining a transformer on ordered multipath sequences transfers to downstream wireless tasks like beam prediction and localization in simulated environments.
-
CENTILE: A Telemetry Foundation Model Evaluated by the Decisions It Drives
One pretrained telemetry model, CENTILE, improves both HPC backfilling and ISP capacity provisioning decisions under replay, with zero-shot transfer across months and domains.
-
Cross-System Neural Precoder: Exploiting Structural Consistency for Fast Adaptation
AO subproblems for digital, analog, and RIS variables share reusable update structure, so a lightweight unfolded network adapts across precoding systems with far less data than a GNN baseline.
-
PERA: A Perceive-Reason-Act Interface Bridging Sensing, Cognitive Reasoning, and Trustworthy Agentic Response for 6G
A three-layer architecture (encoder, learned projector, LLM) enables zero-shot link-state classification and beam prediction with human-readable rationales, outperforming discriminative baselines on the DeepMIMO dataset.
-
Hierarchical Wireless Foundation Model for Multi-Task Optimization
A hierarchical wireless foundation model with a shared channel encoder and prompt-conditioned decoder solves beamforming, scheduling, channel estimation, and beam selection with competitive performance and large laten...
-
Topological sum rule for geometric phases of quantum gates
Geometric phases of a two-qubit gate over a complete basis sum to a multiple of the Hamiltonian winding number, so topology is necessary for entanglement generation.
-
WiFo-2: a generalist foundation model unifies heterogeneous wireless system design
WiFo-2 is a space-time-frequency foundation model pretrained on heterogeneous CSI data that delivers strong zero-shot and few-shot performance across wireless communications and sensing tasks.
-
WiFo-CF: Wireless Foundation Model for CSI Feedback
WiFo-CF is a pretrained wireless foundation model that handles heterogeneous CSI feedback configurations and transfers to localization tasks.
-
LVM4CSI: Enabling Direct Application of Pre-Trained Large Vision Models for Wireless Channel Tasks
A frozen pre-trained vision model can extract wireless channel paths and features, beating conventional estimators in channel estimation and matching specialized networks in sensing with far fewer trainable parameters.
-
CSI2Vec: Towards a Universal CSI Feature Representation for Positioning and Channel Charting
A self-supervised neural network, CSI2Vec, maps wireless channel measurements from different environments and hardware into compact spatial codes that support positioning and channel charting.
-
From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks
HDT-Nets provides a conceptual architecture for 6G networks to coordinate physical AI through holonic digital twins, cognitive value-driven communication, and spatiotemporal integrated information.
-
M3F-UAV: A Missing-Modality Multimodal Foundation Model for Low-Altitude Wireless Sensing
A missing-modality multimodal foundation model fuses RGB, depth, LiDAR, and CSI to support UAV localization, beam prediction, and CSI prediction with graceful degradation when a sensor is missing.
-
Towards channel foundation models (CFMs): Motivations, methodologies and opportunities
A survey and position paper proposing channel foundation models, with experiments on two pretrained CSI models showing gains over a vanilla ViT baseline.
-
Foundation Model-Aided Deep Reinforcement Learning for RIS-Assisted Wireless Communication
A fine-tuned wireless foundation model provides channel embeddings that feed a DDPG agent, which reportedly improves spectral efficiency over DRL with raw CSI and over beam sweeping in DeepMIMO simulation.
-
Foundation Model Empowered Synesthesia of Machines (SoM): AI-native Intelligent Multi-Modal Sensing-Communication Integration
The paper proposes a systematic classification and two roadmaps for using foundation models (LLMs and wireless foundation models) to design Synesthesia of Machines systems for 6G, with preliminary case-study evidence ...
-
WirelessGPT: A Generative Pre-trained Multi-task Learning Framework for Wireless Communication
A pretrained wireless-channel Transformer improves small downstream models for channel estimation, prediction, and activity recognition, and is claimed to support environment reconstruction.
Discussion (0). Continue with ORCID to comment.